Jump to content

Question

Posted

Samsung caught optimizing its phones to score higher on benchmarks

 

It can be tricky figuring out which mobile device to buy. Some get longer battery life than the competition, while others offer more storage, memory, or special features like hands-free operation.

But some folks like to get the model with the fastest processor

Share this post


Link to post
Share on other sites

82 answers to this question

  • 0

Posted

manufacturers have started to optimize their devices specifically to score well on those tests.

 

As they have been for about 5 years now.

 

The Samsung Galaxy S4 smartphone runs the CPU and GPU at full blast whenever you
6 people like this

Share this post


Link to post
Share on other sites
  • 0

Posted

samsung already denied and explained this:

 

http://www.techradar.com/news/phone-and-communications/mobile-phones/samsung-denies-rigging-galaxy-s4-benchmarking-results-1170026

 

 

Which is pretty normal behavior in processors if you ask me, downclock when not in use, upclock when juice is required, how is that cheating?

5 people like this

Share this post


Link to post
Share on other sites
  • 0

Posted

it 's to save battery life, the S4 cpu and gpu doesn't need to run 100% max frequencies for it to the "fastest experience" all the time. Scummy? Really? Come on.... stop leaping to stupid conclusions OP.

3 people like this

Share this post


Link to post
Share on other sites
  • 0

Posted

it 's to save battery life, the S4 cpu and gpu doesn't need to run 100% max frequencies for it to the "fastest experience" all the time. Scummy? Really? Come on.... stop leaping to stupid conclusions OP.

Yeah i was thinking the same thing, I do not see any problem with them cranking the governor on the phone all the way up during a benchmark because that shows what the phone is capable of doing. during normal use we would not want it to be cranked all the way up.

Share this post


Link to post
Share on other sites
  • 0

Posted

it 's to save battery life, the S4 cpu and gpu doesn't need to run 100% max frequencies for it to the "fastest experience" all the time. Scummy? Really? Come on.... stop leaping to stupid conclusions OP.

its what i belive, sorry if it hurts your feelings. I don't know if i am inturpretting it incorrectly or if others are. they way i am reading it is fabricated results. I am sure you also believe what you believe too.

Share this post


Link to post
Share on other sites
  • 0

Posted

its what i belive, sorry if it hurts your feelings. I am sure you also believe what you believe too.

You do realize this happens everywhere?

 

CPUS, GPUs, do you own intel/amd  nvidia or ati?  This is pretty NORMAL BEHAVIOR     Intel used to call this "speedstep" and "powernow and cool and quiet", I think they're both called "turbo something".


http://en.wikipedia.org/wiki/SpeedStep

 

http://en.wikipedia.org/wiki/Intel_Turbo_Boost

 

http://en.wikipedia.org/wiki/PowerNow!

2 people like this

Share this post


Link to post
Share on other sites
  • 0

Posted

benchmarking apps are stupid. real application performance benchmarking is the only way a comparison should be made.

Share this post


Link to post
Share on other sites
  • 0

Posted

I thought they all did it? Nokia was caught out promoting their stabilizer while filming on the move too :P

1 person likes this

Share this post


Link to post
Share on other sites
  • 0

Posted

The issue is not that the CPU/GPU do not run at full speed all the time. Of course they don't.

 

The issue is that the maximum GPU clock even when being pushed by a graphics intensive app like a full screen game is not the same as it is when a benchmark app on the 'booster' list gets launched.

 

That's disingenuous of the phone's maximum real world performance. 

 

Forcing the CPU to full clock speed while a benchmark app is launched also means that the benchmark app doesn't get to see the real world performance of the speed stepping system. If there were delays or problems with the speed stepping you wouldn't be able to see them from the benchmarks because of this trick.

Share this post


Link to post
Share on other sites
  • 0

Posted

the way i am reading it is fabricated results

 

See, actually the device performs at it's absolute best during a benchmark - showing what it is actually capable of achieving. That is what a benchmark is for. The question is not why do this for benchmarks, but rather why not do this for resource intensive games/apps? as the article appears to state the GPU clocks at a lower speed outside of Benchmark apps. - Probably, because some resources are being allocated for the Operating System e.t.c 

Conclusion? The Benchmark represents what the hardware is capable of achieving but does not reflect real life scenarios as less hardware will be available to other apps e.t.c due to resource sharing.

Share this post


Link to post
Share on other sites
  • 0

Posted

You do realize this happens everywhere?

 

CPUS, GPUs, do you own intel/amd  nvidia or ati?  This is pretty NORMAL BEHAVIOR     Intel used to call this "speedstep" and "powernow and cool and quiet", I think they're both called "turbo something".


http://en.wikipedia.org/wiki/SpeedStep

 

http://en.wikipedia.org/wiki/Intel_Turbo_Boost

 

http://en.wikipedia.org/wiki/PowerNow!

from what i read it is not done this way, You may not know what "speed step" and "power now" and "cool and quiet" actually do. Yes they take the actual attainable TOP performance of 100% and knock it down to save power, BUT the difference is the tests that Samsung have rigged the S4 to excel at ARE NOT ATTAINABLE by any program.... that's the error of there results... I think i am reading it correctly.


See, actually the device performs at it's absolute best during a benchmark - showing what it is actually capable of achieving. That is what a benchmark is for. The question is not why do this for benchmarks, but rather why not do this for resource intensive games/apps? as the article appears to state the GPU clocks at a lower speed outside of Benchmark apps. - Probably, because some resources are being allocated for the Operating System e.t.c 

Conclusion? The Benchmark represents what the hardware is capable of achieving but does not reflect real life scenarios as less hardware will be available to other apps e.t.c due to resource sharing.

I could be wrong, but if this top performance is unattainable, then what good is the info? To me its useless data then.

Share this post


Link to post
Share on other sites
  • 0

Posted

I thought they all did it? Nokia was caught out promoting their stabilizer while filming on the move too :p

They do.  HTC has been doing it for years.

Share this post


Link to post
Share on other sites
  • 0

Posted

 


I could be wrong, but if this top performance is unattainable, then what good is the info? To me its useless data then.

from my first response:

 

 

 

In response to the accusation that it has been jimmyingGalaxy S4 benchmark results just to make its handset look good, Samsung says nuh-uh guv'nor.

The company's statement explains that whatever GPU frequencies "BenchmarkBooster" rigged up on theExynos 5 variant are "not intended to improve certain benchmark results" and the same maximum settings can just as easily be used to optimise performance on a users' handset based on what they're using it for.

That's as may be, Samsung, but using phrases like "BenchmarkBooster" in your code is still a bit suspicious in our book. It seems pretty clear that, regardless of whether these maxed out settings have real world use, Samsung was also hoping to beef up its benchmark results.

Now for the science

The full statement reads: "Under ordinary conditions, the Galaxy S4 has been designed to allow a maximum GPU frequency of 533MHz.

"However, the maximum GPU frequency is lowered to 480MHz for certain gaming apps that may cause an overload, when they are used for a prolonged period of time in full-screen mode.

"Meanwhile, a maximum GPU frequency of 533MHz is applicable for running apps that are usually used in full-screen mode, such as the S Browser, Gallery, Camera, Video Player and certain benchmarking apps, which also demand substantial performance.

"The maximum GPU frequencies for the Galaxy S4 have been varied to provide optimal user experience for our customers, and were not intended to improve certain benchmark results. We remain committed to providing our customers with the best possible user experience."

Share this post


Link to post
Share on other sites
  • 0

Posted

To be fair, it is Android.

 

Virtually all the apps are used in full screen unless you have an app that mods things to be windowed.  So why wouldn't that apply to all of them ;)

Share this post


Link to post
Share on other sites
  • 0

Posted

its what i belive, sorry if it hurts your feelings. I don't know if i am inturpretting it incorrectly or if others are. they way i am reading it is fabricated results. I am sure you also believe what you believe too.

 

It's not about beliefs.  The post it just outright nonsense!  But nice try in attempting to make it seem like it matters...

2 people like this

Share this post


Link to post
Share on other sites
  • 0

Posted

from what i read it is not done this way, You may not know what "speed step" and "power now" and "cool and quiet" actually do. Yes they take the actual attainable TOP performance of 100% and knock it down to save power, BUT the difference is the tests that Samsung have rigged the S4 to excel at ARE NOT ATTAINABLE by any program.... that's the error of there results... I think i am reading it correctly.


I could be wrong, but if this top performance is unattainable, then what good is the info? To me its useless data then.

 

I definitely agree.  Performance marks only attainable during benchmarking is not relevant data.

Share this post


Link to post
Share on other sites
  • 0

Posted

its probably not cheating because thats the absolute max it can do, its 480mhz top gpu frequency or whatever is prolly used for thermals and temperature control so it doesnt burn a hole in your hand or the case, or so it doesnt catch fire if your playing games for a long period of time. there will be reasons for keeping it clocked down so its not like they are "cheating". it would prolly be alot worse if its maxed out and get reports of cases burning and melting. its not all a conspiracy.

 

say a PC your cpu and GPU isnt clocked at max frequencys ALL the time, there stepped down but when you do run say a game its clocked to max and as long as your case is well ventilated your fine. stick an i7 w/ a 7970 in a crowded desktop case (not a tower the other one) with cables everywhere, the insides start getting very hot and will destabilize the system cus of the temperature. so downclocking is a necessity sometime

Share this post


Link to post
Share on other sites
  • 0

Posted

I think i am reading it correctly.

 

If you ignore words like "probably" and "seem"...

Share this post


Link to post
Share on other sites
  • 0

Posted

from my first response:

 

Anandtech posted an updated article today to address Samsung's statement:

 

 

 

As we mentioned in our original piece, there's a flag that's set whenever this boost mode is activated: +/sys/class/thermal/thermal_zone0/boost_mode. None of the first party apps get that flag set, only the specific benchmarks we pointed out in the original article. Of those first party apps, S Browser, Gallery and Video Player all top out at a GPU frequency of 266MHz (which makes sense, none of the apps are particularly GPU intensive). I tried running WebGL content in S Browser to confirm (I ran Aquarium with 500 fish), as well as edited some photos in Gallery - 266MHz was the max observed GPU frequency.

 

http://www.anandtech.com/show/7192/update-on-gpu-optimizations-galaxy-s-4

Share this post


Link to post
Share on other sites
  • 0

Posted

at least the phone was smart enough to detect benchmarking behaviors and start pumping up the MHz ...

1 person likes this

Share this post


Link to post
Share on other sites
  • 0

Posted

Sensationalist title. Silly article.

3 people like this

Share this post


Link to post
Share on other sites
  • 0

Posted

What a BS article.. Did Tim Cook write this?

1 person likes this

Share this post


Link to post
Share on other sites
  • 0

Posted

Why would they need to cheat? They pretty much own the market. :D

Share this post


Link to post
Share on other sites
  • 0

Posted

from that same article:

 

 

Isn't this normal behavior (for the camera?) eg. throttle up when needed, throttle down when not?

 

Yes, that is normal behavior. What isn't normal is having a list that calls out specific benchmark apps and tweaks CPU/GPU speeds when those apps are launched.

 

The CPU/GPU clocks should vary like your graph depending on what the benchmark app is trying to do (ie waiting at the menu or doing a benchmark), and peak performance given to the benchmark app should be the same as peak performance given to games and other third party apps.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0

  • Recently Browsing   0 members

    No registered users viewing this page.